Recent evidence suggests that as the market share of managed care increases, expenditures in the fee-for-service sector decrease. This observation demonstrates managed care's extensive effect on the entire health care system and is referred to as a "spillover" effect. It is not known; however, whether quality of care in the fee-for-service sector varies as function of managed care penetration. If lower expenditures in markets with higher penetration result from reductions in the provision of needed services, quality of care in these markets will either decline or improve at a slower rate than in other markets. We propose to examine whether managed care penetration is associated with the quality of care and choice of treatments provided to Medicare patients diagnosed with breast, colorectal, or prostate cancer. We will (1) examine the association between managed care penetration and the quality of care (annual mammography following initial therapy for women diagnosed with early-stage breast cancer; adjuvant chemotherapy for patients diagnosed with stage III colon cancer; adjuvant chemotherapy and radiation therapy for patients diagnosed with stage II or III rectal cancer; recommended surveillance activities (for example, colonoscopy within one year of diagnosis) for patients diagnosed with colorectal cancer; and PSA testing within one year of diagnosis for men diagnosed with localized prostate cancer; (2) examine the association between managed care penetration and the use of treatments that are equally effective but that differ in cost for breast cancer (mastectomy versus breast conserving surgery) and prostate cancer (conservative management versus surgery or radiation therapy); and (3) examine mechanisms by which changes in managed care penetration may lead to changes in patterns of care. That is, whether the influence of managed care on the provider specialty composition within a market and on the availability of technologies within a market will partially explain variations in care associated with changes in managed care penetration. We will use SEER-Medicare data to identify patients diagnosed between 1992 and 1996. We will obtain data from the Health Care Financing Administration and from InterStudy to construct measures of managed care penetration at the county, Health Care Resource Area, and Metropolitan Statistical Area level. We will conduct our analyses within each of these levels using hierarchical regression models. Efforts to reduce cancer- related morbidity and mortality depend upon the provision of services known to be effective. In cases where one effective treatment has not been identified, quality of life can be improved by providing services that are congruent with patients' preferences. This project seeks to determine whether market forces influence the patterns of cancer care. If managed care penetration is negatively associated with the delivery of care, interventions targeted toward policy-makers, providers, and patients will be needed to counteract the effects of market forces.